The concept of diversity has two broad components: richness (also called inventory diversity) and beta-diversity (also called differentiation diversity). Beta-diversity is diversity’s “other component” because it is less studied and more poorly understood than richness. This is partly due to confusion over terminology and concepts of beta-diversity and partly due to lack of data sets that can be used to compute beta-diversity over large areas. In this thesis, I addressed the obstacles to making generalizations about beta-diversity and presented the first empirical analyses of beta-diversity patterns at continental and global scales. In completing this thesis I produced the first maps of beta- diversity at the scale of the Western Hemisphere. While continental and global richness maps are familiar, my maps provide the only look to date at beta-diversity at this scale for multiple classes of terrestrial vertebrates, and offer a striking visual representation of this fundamental component of biological diversity.
In my review chapter, I traced the conceptual history of beta-diversity from its origins in community ecology to its many applications today. My review revealed the historical development of a single phenomenon of differentiation diversity, the compositional change in species regardless of scale or mechanism. I argued that this unifying concept ties together the various divisions previous authors have made. I further recommended that the term beta-diversity continue to be applied to this phenomenon because of the widespread use and the historical roots of this concept.
The three empirical chapters of this dissertation examined broad-scale beta-diversity of terrestrial vertebrates across multiple locations and extents, with the underlying purpose of determining whether
certain generalities exist. The taxonomic and geographic scope of my analyses of three terrestrial vertebrate classes (birds, mammals, and amphibians), 7,667 terrestrial vertebrate species across the Western Hemisphere and 5,817 amphibian species globally, exceeds that of any previous analyses of beta-diversity. I used the Western Hemisphere data, at a grain of 100 km x 100 km, to perform a thorough assessment of cross-taxon congruence in broad-scale beta-diversity and of the relationship between beta-diversity and species richness. With the global amphibian data, I conducted a thorough analysis of the distance decay relationship across biomes and biogeographic realms. These analyses used biogeographic realms as a surrogate for differences in history, biomes as a surrogate for differences in environment, and topographic complexity as a surrogate for local environmental heterogeneity and dispersal barriers. Similarly, the three vertebrate classes represented distinct taxonomic groups with different ranges of life history and dispersal abilities, and which can be expected to represent differences in historical patterns of evolution. While these are coarse surrogates of underlying mechanisms, they were used to explore basic broad-scale patterns in beta-diversity.
I addressed three general expectations for broad-scale beta-diversity: Are beta-diversity patterns congruent across taxa? Is there a relationship between beta-diversity and species richness (measured at the same scale)? Is there systematic variation in beta-diversity across biomes and biogeographic realms?
Many of my results from analyses at the Western Hemisphere extent were consistent for the three vertebrate classes. I also found strong patterns in global amphibian beta-diversity across certain contrasting sets of biomes and of realms. However, my results also showed interesting variation with differences in spatial extent and geographic location, between taxa, between the high and low extremes of diversity components, and between metrics (i.e., between initial similarity and distance decay rate, and between Jaccard’s index of similarity and an index designed to remove the effect of species richness gradients).
In the following paragraphs I summarize the findings that support generalizations in cross-taxon congruence in beta-diversity, the relationship of beta-diversity to species richness, and systematic
variation in beta-diversity across biomes and biogeographic realms. I then outline the variations I found relating to each of these questions. Lastly, I make recommendations for future research.
Generalities in Broad-Scale Beta-Diversity
Are Beta-Diversity Patterns Congruent Across Taxa?
Amphibian, bird, and mammal beta-diversity patterns were largely congruent across the extent of the Western Hemisphere. There was a striking association of high beta-diversity and mountains apparent when mapped, which was also confirmed statistically for each taxon. My work showed an interesting discrepancy between congruence in highest beta-diversity areas, which were similarly distributed for the three taxa, and congruence in lowest beta-diversity areas, which were largely distinct. This suggests that similar processes lead to high levels of differentiation of these taxa, but the ecological and biogeographic factors influencing low levels of differentiation vary.
Is There a Relationship Between Beta-Diversity and Species Richness?
Beta-diversity and richness exhibited disparate patterns for all three taxa at the scale of the Western Hemisphere. For each taxon, there was considerable spatial segregation between the highest areas of the two diversity components. This demonstrates that patterns of beta-diversity contain information that cannot be provided by measures of species richness alone, and has implications for the mechanisms underlying broad-scale diversity patterns. That not all species-rich, tropical areas have rapid species turnover suggests that the role beta-diversity plays producing the high richness of the tropics is not straightforward, at least at the scales I measured.
The importance of mountains was apparent from my analyses of both cross-taxon congruence in beta-diversity and the relationship between beta-diversity and richness. As noted above, beta-diversity for all three taxa was high in mountainous areas. Moreover, the areas of highest beta-diversity for the three groups overlapped in the Northern Andes. Mountains were also at the intersection of the highest beta-diversity and richness areas for each taxon. That high levels of the two diversity components co- occur in topographically complex areas both within and outside of the tropics confirms the importance of history and topography in generating diversity.
Is There Systematic Variation in Beta-Diversity Across Biomes and Biogeographic Realms?
Variation in amphibian initial similarity and distance decay rates was complex, yet I found evidence suggesting that strong patterns may exist between certain broad climate contrasts or for particular historical differences. For instance, initial similarity within biomes in the Australasian realm were generally much lower than within their counterparts in other realms, and distance decay rates within Afrotropical biomes were more rapid than within the same biomes in the Neotropics. Another interesting trend was that distance decay rates in grasslands were more rapid than in forests within both temperate and tropical regions.
Such trends suggest that while biogeographic history and environment are both important in influencing the degree of change in species composition at near and far distances, the relative strength of each in determining differences in beta-diversity may be contingent upon particular aspects of climate or biogeographic history. Determining the mechanisms driving these differences, however, will require more detailed analyses. Comparing my findings for amphibians with distance decay rates for other taxa will help ascertain whether these results are indicative of a widespread trend or are particular to amphibians.
The preceding paragraphs have described, for three questions concerning broad-scale beta- diversity, the results I found to be general across taxa or regions. Below, I discuss the results relating to each of these questions which varied between taxa, geographic location and spatial extent, or metric used to measure beta-diversity.
Variations in Broad-Scale Beta-Diversity
Beta-Diversity Congruence
When measured at extents smaller than the Western Hemisphere, the strength of congruence in beta-diversity varied with geographic location and spatial extent, as well as between pairs of taxa. For example, each pair of taxa showed much stronger congruence within the Neotropical realm than within the Nearctic realm. Correlations measured at the same extent and location also differed between taxonomic pairs. Within the Nearctic realm, for instance, amphibian and mammal beta-
diversity showed a moderate degree of congruence, as did bird and mammal beta-diversity, but beta- diversity of amphibians and birds showed no significant congruence. The stronger congruence in beta-diversity within the Neotropics compared to the Nearctic is consistent with the historical differences between these realms. The Neotropics have experienced greater historical stability than the Nearctic, which had more severe climatic fluctuations and more extensive glaciation during the Pleistocene. Future research is needed to determine whether this pattern extends beyond the Western Hemisphere, for example, whether the high congruence in the Neotropics is also found in other tropical realms such as the Afrotropics, and how the level of congruence found in the Nearctic compares to the level in the Palearctic.
Relationship Between Beta-Diversity and Species Richness
For each taxon, the strength of the correlation between beta-diversity and richness, and whether the relationship was positive or negative, varied between biogeographic realms. Mammalian beta- diversity and richness, for example, were positively correlated in the Nearctic, but negatively correlated in the Neotropics. The results within one biogeographic realm also varied between taxa. For instance, in contrast to the positive correlation for mammals in the Nearctic, bird beta-diversity and richness in that realm had a weak negative correlation.
Variation in Beta-Diversity Across Biomes and Realms
My global analysis of distance decay relationships of amphibians revealed several interesting contrasts between the two distance decay parameters: initial similarity level and distance decay rate. For instance, there was a strong relationship between the topographic complexity in a region and initial similarity level, but there was no significant relationship between topographic complexity and distance decay rate. Moreover, the rate of distance decay measured for a region, and the variation in distance decay rates between regions, were much less affected by factoring out local richness gradients than were the level of initial similarity measured for a region and the variation in initial similarity between regions.
Future Research
My results provided support for the existence of general trends in beta-diversity at broad scales, at least for terrestrial vertebrates. Below I outline four areas of research which are either important to elucidating generalities about broad-scale beta-diversity or important to the practical application of beta-diversity to conservation.
Taxonomic and Geographic Scope
For the vertebrate classes I examined, cross-taxon congruence and the relationship between beta- diversity and richness should be examined on other continents and at different grains, and the global variation in distance decay should be investigated for birds and mammals. To be able to state with certainty that there are general trends in beta-diversity, however, the scope of study must be expanded beyond terrestrial vertebrates, with emphasis on analyzing beta-diversity patterns across a wide phylogenetic/taxonomic range of organisms. The marine and freshwater realms offer exciting prospects for testing beta-diversity generalities. Freshwater systems are naturally isolated, while marine systems are seemingly open, and the processes relating to dispersal are so different than in terrestrial systems (at least from the perspective of a terrestrial ecologist!). Although beta-diversity in marine and freshwater systems is studied at many scales, the vast majority of beta-diversity studies, including those at broad-scales, have been terrestrial.
Environmental and Historical Processes
In addition to increasing the taxonomic and geographic coverage of analysis, future research should include more detailed examination of the processes underlying broad-scale beta-diversity. In particular, understanding the relative influence of environment and history, and under what circumstances one is more important than the other, is central to determining whether or not underlying trends exist across taxa and regions; it also has practical application for surrogate methods in conservation planning.
The analyses in this dissertation were all based on presence-absence data, which may produce different patterns than abundance data. In contrast with the community ecology roots of beta- diversity, beta-diversity studies at large scales are generally based on data sources limited to presence-absence data, such as range maps. As the availability of abundance data at large scales increases, studies comparing the two data types will provide another perspective on broad-scale beta- diversity. Another interesting avenue for measures of beta-diversity is a development of a metric that incorporates phylogenetic dissimilarity as well, so that more distantly related species counts as more distantly than more closely related species.
The Role of Beta-Diversity in Conservation Planning
Although beta-diversity has received indirect attention in the conservation literature for some time, notably in the early SLOSS debates (e.g., Simberloff & Abele 1976) and in the more recent profusion of complementarity algorithms (Sarkar 2006), few methods for directly addressing beta- diversity in conservation planning have been developed. Complementarity is an important principle for designing representative conservation networks, but beta-diversity has many benefits for conservation apart from its link with this principle. For example, beta-diversity analyses can help identify areas where species face increasing threat to persistence, because beta-diversity is often high where species’ ranges are particularly susceptible to climatic variability such as steep environmental gradients and centers of endemism, or regions where successful conservation strategies may be resource intensive, because gradients of rapid species turnover will require closely spaced protected areas in order to effectively conserve biodiversity.
Fortunately, there is increasing interest in integrating beta-diversity into systematic conservation planning. One example is the development of conservation surrogates based on modeling compositional dissimilarity, which can improve biodiversity representation for data poor regions (Ferrier 2002; Ferrier et al. 2004; Steinitz et al. 2005). Several recent studies have incorporated turnover measures into area selection algorithms, with the goal of addressing persistence (Fairbanks
of determining appropriate reserve spacing has also received recent attention (Wiersma & Urban 2005). These are all important issues in need of more in-depth research.
References
Fairbanks, D.H.K., Reyers, B. & van Jaarsveld, A.S. (2001) Species and environment representation: selecting reserves for the retention of avian diversity in KwaZulu-Natal, South Africa. Biological Conservation 98, 365-379.
Ferrier, S. (2002) Mapping spatial pattern in biodiversity for regional conservation planning: Where to from here? Systematic Biology 51, 331-363.
Ferrier, S., Powell, G.V.N., Richardson, K.S., Manion, G., Overton, J.M., Allnutt, T.F., Cameron, S.E., Mantle, K., Burgess, N.D., Faith, D.P., Lamoreux, J.F., Kier, G., Hijmans, R.J., Funk, V.A., Cassis, G.A., Fisher, B.L., Flemons, P., Lees, D., Lovett, J.C. & Rompaey, R.S.A.R. (2004) Mapping more of terrestrial biodiversity for global conservation assessment. BioScience 54, 1101-1109.
Sarkar, S. (2006) Ecological diversity and biodiversity as concepts for conservation planning: comments on Ricotta. Acta Biotheoretica 54, 133-140.
Simberloff, D.S. & Abele, L.G. (1976) Island biogeography theory and conservation practice. Science
191, 285-286.
Steinitz, O., Heller, J., Tsoar, A., Rotem, D. & Kadmon, R. (2005) Predicting regional patterns of similarity in species composition for conservation planning. Conservation Biology 19, 1978- 1988.
Wiersma, Y.E. & Urban, D.L. (2005) Beta diversity and nature reserve system design in the Yukon, Canada. Conservation Biology 19, 1262-1272.